Table 3 AUC and its standard deviation (STD), specificity of Task 4: pure atypia outcome vs. upstaged atypia outcome. 10-fold cross-validation was used for model training
From: Deep learning of mammogram images to reduce unnecessary breast biopsies: a preliminary study
Task | View | AUC (STD) | Specificity (%) | ||
|---|---|---|---|---|---|
Given disease sensitivity = 100% | Given disease sensitivity = 99% | Given disease sensitivity = 95% | |||
Task4 | CC | 0.67 (0.14) | 35 | ||
MLO | 0.58 (0.16) | 18 | |||
CC + MLO | 0.65 (0.14) | 18 |